Last Active
  • Adding water cooling to the Mac Studio does surprisingly little

    Experienced PC builders will be familiar with the benefits of water cooling, including the potential to offer better thermal conditions than air-based methods, the possibility of improved chip performance, and a reduction in noise.
    Reminder that the "water cooling" here is basically like how "wireless charging" is used to describe induction charging. Using the words "water cooling" is conveying more than what it is actually doing. Ultimately, these cooler designs are transferring heat from the chip to the ambient atmosphere. Virtually all them do it by blowing air across radiator fins (heat sink in my parlance). This is an "air-based" method.

    Basically every single "water cooler" is a closed loop that pumps water from the chip block to a radiator and a fan is used to blow air across it to transfer heat to the ambient air. So if there is a Venn diagram, PC water coolers would be a circle inside an "air-based" cooler.

    Moreover, all of the high performance "air-based" coolers, as you are thinking of them, like what is inside the Mac Studio or the MacBook Pro or Mac Pro, are "liquid coolers". They employ a heat pipe which employs a fluid medium that transfers heat from the chip block to the radiator fins. They might employ a 2 phase loop, or just be single phase, but it's a pipe with a fluid inside. The fluid transfers heat from the chip block to the radiator fins, just like what a water cooler does.

    The one big advantage a water cooler has is that it uses flexible hosing. Note that I didn't say heat transfer performance. Flexible hosing enables a PC builder to do neat stuff in their case in an affordable way. The heat pipes in "air-based" coolers are basically inflexible and are custom designed for the application. You get what you get from the vendor. "Heat pipes" are as efficient if not more efficient in transferring heat than pumped water, I think. So the heat transfer properties of pumped water is not a heat transfer advantage. Flexible hosing is nice though, as PC cases have a lot of variation. PCs often have limited volume or not-optimal volumes for heatsinks or radiators. The flexible hosing enables the radiator to be 6", 12", 18" long with more advantageous fan setups and radiator locations. This is good for the PC market where there is a lot of case and motherboard variation.

    If you start from a blank page, you can do everything with what you think of as air-based coolers. Just look at the Mac Pro.

  • Emergency SOS via satellite expanding to more countries in 2022, sketchy report says

    The article says "The satellite connection feature can also be used to send the user's location to friends or family members." This suggests that the user's location is not necessarily sent to the satellite. And if it is not sent to the satellite, then how would the satellite know the location of the phone? And if it doesn't know the location of the phone, how does it prevent callers from, say, Mexico or Cuba, (which are near the US) from using this feature? I haven't seen anyone asking if iOS uses geo-fencing to restrict users to certain countries.
    The phone has GPS. It will work anywhere around the Earth as long as the GPS antenna can get a lock on 3 or 4 GPS satellites. So, GPS coordinates of an iPhone are transmitted to a GlobalStar satellite, it transmits them to GlobalStar relay stations, the relay station than routes it to cellular network and or to emergency services of the supported country.

    GlobalStar satellites won't know where the phones are. It only knows that it has established a connection to a device, the device has sent it the destination of where it wants the message to go. It may not even know that. It might only be the relay station that is capable of figuring where a message needs to go or where a call needs to go. The content of the message? It's going to be encrypted if it isn't already.

    A user can connect to a GlobalStar satellite anywhere on the planet with line of sight to the satellite. It will work in China, in Cuba, in the middle of ocean. However, would a relay station be able to route an SOS to a country's emergency services? Or a client device within a certain country? Well, it depends on whether the country will allow the relay station to connect to its telecomm services in the country. Hence, Apple has to coordinate that. GPS coordinates routed to Find My? Yes, that should be supported anywhere on the planet with GlobalStar satellites in line of sight.
  • Supply chain gearing up to ship new MacBook Pro

    blastdoor said:
    While the notion of 3nm M2 Pro/Max now seems highly unlikely, I wonder if they might be fabbed on 4nm (like the A16)?

    TSMC's 4nm process is a refinement of the 5nm. I believe it uses the same design rules as 5nm, which means that Apple need not change the design of the CPU and GPU cores in the 5nm M2 for use in a 4nm M2 Pro/Max. 
    I don’t think there is much difference between N5P and N4. Apple will be able to eek out 10% to 20% performance gains from either just as they did between M1 to M2 or A15 to A16. 

    Apple does need to fix the poor GPU core scaling in their SoCs. Perhaps it is a memory controller issue, whatever it is, it needs to be fixed. That’s probably the number 1 thing they needed to have worked on over the past 2 years. 
  • iOS 16's Clean Energy Charging arriving later in 2022

     It is unclear exactly how iOS would be able to know when "cleaner energy sources" are being used, but it is likely checking in with Apple's servers to determine the grid's status. It will also be limited only to the United States, which suggests there is some data sourcing at play.
    Apple just asks the grid operator for their realtime power mix and their future power prediction, by type. The rest is having the phone calling their server for optimal charging times. The data is public and you can check it yourself, of if you are good at HTML parsing, you can continuously crawl for the data if wanted.
    The grid operator definitely knows what power source is energizing the grid on an hour by hour basis, at least, if not minute by minute basis. If you live in Texas ERCOT, the best time to charge is from 2 am to 6 am or so, when wind power is about 20% to 50% of grid, on average and depending on time of year. In a few years, it's going to be something like 30% to 70% wind at those hours, with solar+wind creeping towards 20% to 50% during the day.

  • Apple introduces iPhone 14 & iPhone 14 Plus -- with satellite connectivity

    sirbryan said:
    Many commenters seem to assume that satellite signals can be had anywhere, anytime, globally.

    Typically communications satellites are geosynchronous; they appear stationary to us from the ground. This way the uplink stations don't have to have directional tracking antennas. Most TV, telephone, and Internet systems are this way (think Dish Network, DirecTV, HugesNet, ViaSat, etc.).  Even my Starlink dish stays pointing in one direction, even though their satellites are much closer in Low Earth Orbit.

    If you don't have a bird (satellite) covering a particular country, that service won't work in that place.

    Some providers may have birds moving in non-geosynchronous orbits (i.e. north-south or at various sloping angles, faster or slower than the earth spins, etc.), so in those cases, you'd have a store-and-forward messaging system. SpaceX alluded to this being the case for how their collaboration with T-Mobile might work. In that case, while their service could technically work anywhere in the world, they're beholden to regulatory bodies when transmitting on licensed frequencies in International air space.  And again, the birds need to communicate with the ground at some point to pass the messages on.
    This type of service, satellite phone service, is provided by a 50 to 100+ satellites in a low Earth orbit. GlobalStar, Irridium, et al. GPS, GLONASS same way but their signaling is one-way. About 200 to 500 miles above the surface of the Earth, in a multitude of orbits from polar to equatorial. Geosynchronous satellites are 22,000 miles above the surface of the Earth in equatorial orbits. 22k miles will be a challenge for something the size of a phone. 200 to 2000 miles, much more reasonable for something the size of a cell phone, which is amply demonstrated by the satellite phones in existence for 2 decades now.

    The key regulatory bottleneck for this type of service will be in the ground relay stations that take the satellite signals and relay to a mobile service and go to emergency services. Notionally, a government can say that the satellite provider is prohibited from using the spectrum utilized by the service, but that is like saying people are prohibited from using light. They can regulate the relay stations and the relay spectrum, so a message won't be serviced inside the country. That message however will go to wherever their are relay stations and forwarded to the destination/client.

    I'm still unsure of a message going between satellite clients: sat phone to sat phone. Apple will have to pay for such a service from the satellite provider, so, it isn't going to be free. I can see SOS and Find My being free for iPhone buyers, but 2-way comm, no way.